--- title: "agentdojo vs baseline-defenses" type: "comparison" canonical_url: "https://www.graphcanon.com/compare/ethz-spylab-agentdojo-vs-neelsjain-baseline-defenses" tools: ["ethz-spylab-agentdojo", "neelsjain-baseline-defenses"] --- # agentdojo vs baseline-defenses *GraphCanon updated Aug 5, 2026* ## Verdict Pick agentdojo if agentDojo serves as a benchmarking environment to evaluate security attacks, like prompt injection, and defenses for Large Language Model (LLM) agents; pick baseline-defenses if a toolkit for evaluating defenses against adversarial attacks on aligned language models, focusing on perplexity filter and paraphrase defense strategies. [agentdojo](https://agentdojo.spylab.ai/) reports 716 GitHub stars, 188 forks, and 41 open issues, last pushed Jun 2, 2026. [baseline-defenses](https://github.com/neelsjain/baseline-defenses) has 34 stars, 1 forks, and 0 open issues, last pushed Oct 26, 2023. Figures are from public GitHub metadata via [agentdojo's repository](https://github.com/ethz-spylab/agentdojo) and [baseline-defenses's repository](https://github.com/neelsjain/baseline-defenses). | | [agentdojo](/tools/ethz-spylab-agentdojo.md) | [baseline-defenses](/tools/neelsjain-baseline-defenses.md) | | --- | --- | --- | | Tagline | A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents | Research code for evaluating defenses against adversarial attacks on aligned language models | | Stars | 716 | 34 | | Forks | 188 | 1 | | Open issues | 41 | 0 | | Language | Python | Python | | Adopt for | AgentDojo serves as a benchmarking environment to evaluate security attacks, like prompt injection, and defenses for Large Language Model (LLM) agents. | A toolkit for evaluating defenses against adversarial attacks on aligned language models, focusing on perplexity filter and paraphrase defense strategies. | | Persona | - | - | | Runtime | - | - | | License | MIT | - | | Categories | AI Agents, Evaluation & Observability | Evaluation & Observability | ## Trust and health _Sourced signals - not a safety guarantee. No winner column._ | | [agentdojo](/tools/ethz-spylab-agentdojo.md) | [baseline-defenses](/tools/neelsjain-baseline-defenses.md) | | --- | --- | --- | | Maintenance | Steady (60%) | Dormant (18%) | | Days since push | 63d | 1013d | | Open issues (now) | 41 | 0 | | Owner type | Organization | User | | Full report | [trust report](/tools/ethz-spylab-agentdojo/trust.md) | [trust report](/tools/neelsjain-baseline-defenses/trust.md) | ## Decision facts: agentdojo - **Pricing:** freemium - Open-source under the MIT License. Some advanced features might require additional libraries or APIs. - **Requirements:** Min 8 GB RAM - **Adopt for:** AgentDojo serves as a benchmarking environment to evaluate security attacks, like prompt injection, and defenses for Large Language Model (LLM) agents. ## Decision facts: baseline-defenses - **Adopt for:** A toolkit for evaluating defenses against adversarial attacks on aligned language models, focusing on perplexity filter and paraphrase defense strategies. ## Choose when ### Choose agentdojo if… - Pricing: Open-source under the MIT License. Some advanced features might require additional libraries or APIs.. - Requirements: Min 8 GB RAM. - Tags unique to agentdojo: benchmark, large language models, prompt-injection, security. - Also covers AI Agents. - AgentDojo serves as a benchmarking environment to evaluate security attacks, like prompt injection, and defenses for Large Language Model (LLM) agents. ### Choose baseline-defenses if… - Tags unique to baseline-defenses: adversarial-attacks, defense strategies, paraphrase defense, perplexity filter. - - When you need to evaluate the effectiveness of baseline defenses such as the perplexity filter or paraphrase defense in protecting aligned language models from adversarial attacks. - Leaner open-issue backlog (0). ## When NOT to use agentdojo - AI Agents: Don't use an agent loop when a deterministic workflow would do; agents add latency, cost, and non-determinism. - Evaluation & Observability: Defer heavyweight eval infra only until you have real traffic - never skip it once users depend on answers. ## When NOT to use baseline-defenses - - Do not use if you require comprehensive coverage of all possible defensive measures. This tool specifically lacks detailed code for retokenization defenses involving BPE-dropout. - - If your scenario demands more advanced or specialized defense mechanisms beyond the scope of baseline strategies, this repository will fall short on delivering those. ## Common questions ### What is the difference between agentdojo and baseline-defenses? agentdojo: A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents. baseline-defenses: Research code for evaluating defenses against adversarial attacks on aligned language models. See the comparison table for live GitHub stats and shared categories. ### When should I choose agentdojo over baseline-defenses? Choose agentdojo over baseline-defenses when Pricing: Open-source under the MIT License. Some advanced features might require additional libraries or APIs.; Requirements: Min 8 GB RAM; Tags unique to agentdojo: benchmark, large language models, prompt-injection, security; Also covers AI Agents; AgentDojo serves as a benchmarking environment to evaluate security attacks, like prompt injection, and defenses for Large Language Model (LLM) agents. ### When should I choose baseline-defenses over agentdojo? Choose baseline-defenses over agentdojo when Tags unique to baseline-defenses: adversarial-attacks, defense strategies, paraphrase defense, perplexity filter; - When you need to evaluate the effectiveness of baseline defenses such as the perplexity filter or paraphrase defense in protecting aligned language models from adversarial attacks; Leaner open-issue backlog (0). ### When should I avoid agentdojo? AI Agents: Don't use an agent loop when a deterministic workflow would do; agents add latency, cost, and non-determinism. Evaluation & Observability: Defer heavyweight eval infra only until you have real traffic - never skip it once users depend on answers. ### When should I avoid baseline-defenses? - Do not use if you require comprehensive coverage of all possible defensive measures. This tool specifically lacks detailed code for retokenization defenses involving BPE-dropout. - If your scenario demands more advanced or specialized defense mechanisms beyond the scope of baseline strategies, this repository will fall short on delivering those. ### Is agentdojo or baseline-defenses more popular on GitHub? agentdojo has more GitHub stars (716 vs 34). Stars measure visibility, not whether either tool fits your constraints. ### Are agentdojo and baseline-defenses open source? Yes - both are open-source projects on GitHub. ### Where can I find alternatives to agentdojo or baseline-defenses? GraphCanon lists graph-backed alternatives at [agentdojo alternatives](/tools/ethz-spylab-agentdojo/alternatives) and [baseline-defenses alternatives](/tools/neelsjain-baseline-defenses/alternatives) ([agentdojo markdown twin](/tools/ethz-spylab-agentdojo/alternatives.md), [baseline-defenses markdown twin](/tools/neelsjain-baseline-defenses/alternatives.md)), ranked by typed relationship edges rather than popularity votes. ### Is there a machine-readable version of this comparison? Yes. The markdown twin at [this comparison](/compare/ethz-spylab-agentdojo-vs-neelsjain-baseline-defenses.md) mirrors this page for agents and LLM crawlers, with the same stats table and FAQ answers. ### Which is better maintained, agentdojo or baseline-defenses? agentdojo: Steady. baseline-defenses: Dormant. Compare maintenance labels, days since push, and release cadence in the trust section below - stars alone do not measure maintenance. ### Where are the full trust reports for agentdojo and baseline-defenses? GraphCanon publishes per-repo trust reports with dated maintenance, provenance, and scan summaries: [agentdojo trust report](/tools/ethz-spylab-agentdojo/trust); [baseline-defenses trust report](/tools/neelsjain-baseline-defenses/trust). --- **Machine-readable endpoints** - JSON: [`/api/graphcanon/graph?tool=ethz-spylab-agentdojo`](/api/graphcanon/graph?tool=ethz-spylab-agentdojo) - LLM index: [/llms.txt](/llms.txt) - Full corpus: [/llms-full.txt](/llms-full.txt) _GraphCanon - The knowledge graph for AI development. https://www.graphcanon.com/_